LANGUAGE CHOICE IN MULTILINGUAL COMMUNITIES-THE CASE OF
LARTEH, GHANA
Mercy Akrofi Ansah1
Abstract
In a multilingual community, the multilingual speaker needs to make the right language
choice which principally depends on the domain of usage and the linguistic repertoire of
speech participants. This paper investigates factors that govern language choices that
multilingual speakers make in Larteh, a multilingual community. The study is informed by
insights from the Markedness Model, developed by Myers-Scotton (1993; 1998). Larteh is a
non-reciprocal bilingual community: Leteh2-Akuapem Twi (Johnson 1973: i) and English is
the third language for those who have had formal education. In this paper, three domains of
language use are examined: education, tradition and religion. Data from an interview survey
on language use and participant observations are employed. The paper notes that due to
changes in various spheres of life in Larteh, current language use patterns in the community
differ from what pertained about three decades ago (Johnson 1973, 1975). Subsequently,
factors that determine language choice are gradually undergoing some modification.
1.0 Introduction
This paper explores factors that inform language choices that are made by multilingual
speakers in Larteh. People who speak more than two languages are often confronted with
making the right language choice within a particular domain. Language choice is informed by
the kind of participants in a communication situation, the topic, social distance and also
location. Studies on language use in multilingual communities in Africa abound (see Yakubu
et al 2012; Kamwangamalu 2000; Ncoko et al 2000; Bodomo et al 2009 among others).
Brokensha (1966) and Johnson (1973, 1975) constitute significant research on language use
in Larteh. Whereas Brokensha (1966), an ethnographic study, briefly discussed the subject of
language use in Larteh in his introductory chapter, those by Johnson are thorough studies of
the subject matter. In his introductory chapter, Brokensha presented an overview of the
linguistic situation of Larteh, and also the functional distribution of the three major
languages: Leteh, Twi and English. Leteh was described as a domestic language, which was
learnt by children as their first language. English was learnt at school and school lessons were
taught in Twi. To some extent, Twi was considered to be a prestigious language, and
associated with education and Christianity. In church, English was used out of necessity,
especially when there were non-Africans in the congregation. Brokensha further commented
that language used in church was Twi even when the congregation was made up of only
Leteh speakers. In this regard, he cited the example of the Pentecostal church at Larteh where
all the worshipers were Leteh speakers and yet, Twi was used. Besides the use of English in
school, it was spoken by educated adults in conversation. School children who wished to
practise their oral skills or impress non-literates also used English outside the classroom
1
Mercy Akrofi Ansah is a Research Fellow at the Institute of African Studies, University of Ghana
In this paper, Leteh is the spelling adopted for the name of the language of the people of Larteh. Elsewhere in
the literature, the language is also known as Larteh or Lɛtɛ.
2
setting. The present study will verify if the functional distribution of the three languages as
described by Brokensha still persists in Larteh.
Johnson (1973) was an in-depth study of the patterns of language use in Larteh, as an
example of a bilingual community. The work described all aspects of language use in the
Larteh community, and his findings corroborated those by Brokensha (1966). Johnson further
stressed the domestic function of Leteh by commenting that although school children spoke
Twi and English for practice at home, it was often considered rude and disloyal, especially
when spoken in the presence of adults. The author commented that the home was essentially
a monolingual environment at Larteh. With regard to language choice at church, the author
further noted that Twi was the language of the Bible, hymnals and prayer books. English was
used only when visitors were present, as was the case when the author and his wife visited the
Larteh Presbyterian Church as part of his field trip.
With regard to education, the Basel Missionaries supported the use of Twi and made
it the foundation of their educational system. Twi was the medium of instruction, and Larteh
children only learnt it at school. Some attempts were therefore made by the Gold Coast
educational authorities to post teachers who could speak Leteh or Kyerepong (Guan
language) to Larteh to teach the Lower Primary classes. In school, the use of Leteh was
permitted but not encouraged. Codeswitching between Leteh-Twi was also not encouraged as
it was believed to have the potential of impeding the learning progress of school children.
During traditional ceremonies, Leteh was the primary choice. However, in some instances of
libation pouring, orations and speeches, some Twi phrases were heard. In these instances,
English was not used except among the audience. Finally, Johnson (1975) discussed the
triglossic relationship among Leteh, Twi and English. He described the language situation as
one characterized by a division of communicative functions among the three languages, a
situation which is the result of a gradual change over a long period. The author asserted that
the functional distribution of languages in Larteh was not random, and proposed a set of rules
which could be used to predict language choice in a multilingual community such as Larteh.
These were repertoire rules, situational rules and metaphorical rules.
Changes in the various spheres of life in a community are very likely to affect
language use. Broad areas which are susceptible to change include education, technology and
social. Undoubtedly, conditions which prevailed during the period of previous researches
(Brokensha (1966); Johnson (1973, 1975) have changed, hence transformations in the
patterns of language use in Larteh are to be expected. It is therefore in the light of these
changes that this study is undertaken. This paper will therefore investigate factors that impact
language selection in Larteh in the light of these changes. The paper is organized as follows:
after the introductory section, sources of data and collection methods are discussed in section
two. Myers-Scotton’s Markedness Model (1993; 1998) and its application to the study are
dealt with in section three. In section four, a sociolinguistic background of Larteh is given. In
the three sections that follow, patterns of language choice in the three domains of language
usage are discussed. The discussion is summarized, and conclusions are drawn in section
eight.
2.0 Data and Methods
An interdisciplinary method of data collection was used for the study2; linguistic and
anthropological. A linguistic survey of Larteh (LISL) was conducted to collect information
on individual and societal linguistic repertoires. The survey was conducted from September 5
- September 9, 2011 with 12 Research Assistants: 6 locals, paired with 6 National Service
personnel from the University of Ghana. One Research Assistant helped with data analyses.
Results of the survey also gave indication of patterns of language use in Larteh. In all, 418
interviews were conducted to specifically determine language types that are used in the
community. This number represented about 5% of the total population of Larteh - 8,310
(Ghana Housing and Population Census, 2000). The interviewees who were randomly
selected were aged 10 years and above. Furthermore, Primary 1 classroom teaching sessions
in three public schools in Larteh were observed. The schools were the Anglican Primary
School on September 12, 2011; the Presbyterian Primary School on September 14, 2011; and
the Methodist Primary School on September 15, 2011. Funeral activities and Church services
were also observed. Two funeral sessions were observed on October 29, 2011 and November
26, 2011. With regard to language choice in the church, notes taken at the Presbyterian
Church and Lighthouse Chapel on January 22, 2012 and February 5, 2012 respectively are
used for illustration. These exercises enabled the researcher to investigate the choices that
teachers, family spokesman and church leaders make in multilingual contexts such as the
classroom, funeral gatherings and church services.
3.0 Theoretical Framework
This study is done within the framework of Myers-Scotton’s Markedness Model
(1993; 1998). Markedness, according to Myers-Scotton (1998: 4), relates to the choice of one
linguistic variety over other possible varieties. According to Myers-Scotton (1993: 84), each
language in a multilingual community is associated with particular social roles, which she
calls rights-and-obligations (RO) sets. The speaker-hearer signals her understanding of the
current situation by deciding to speak a particular language. The choice of a particular
language is also an indication of her relevant role within the context.
The Markedness Model is stated in the form of a principle, the negotiation principle, and
three maxims. The negotiation principle presents the theory’s central idea: “Choose the form
of your conversational contribution such that it indexes the set of rights and obligations which
you wish to be in force between the speaker and addressee for the current exchange (MyersScotton 1993: 113).
Three maxims emerge from this principle: the unmarked choice maxim; the marked
choice maxim and the exploratory choice maxim. The unmarked choice maxim states, “Make
your code choice the unmarked index of the unmarked RO set in talk exchanges when you
wish to establish or affirm that RO set” (Myers-Scotton 1993: 114). The marked choice
maxim directs, “Make a marked code choice…when you wish to establish a new RO set as
unmarked for the current exchange” (Myers-Scotton 1993: 131). Finally, the exploratory
choice maxim states, “When an unmarked choice is not clear, use CS (codeswitching) to
2
This research was funded by the American Council for Learned Societies (ACLS), under the African
Humanities Programme (AHP).
make alternate exploratory choices as candidates for an unmarked choice and thereby as an
index of an RO set which you favor” (Myers-Scotton 1993: 142). In this sense, social
meanings of language choice, as well as the causes of alternation are defined completely in
terms of participant rights and obligations.
The Markedness Model uses the marked versus unmarked distinction to explain the
social and psychological motivations for making one language choice over another. What
community norms would predict is unmarked; what the community norms would not predict
is marked (Myers-Scotton 1998: 5). As Kieswetter (1995: 15) explains, the unmarked choice
is considered as the expected choice within that particular context, whereas making a marked
choice often carries extra social meaning. The Markedness Model (1993; 1998) is appropriate
for the present study as it relates to the choice of one linguistic variety over other possible
varieties in a speech context. In this paper, marked and unmarked choices that speech
participants make in the education, tradition and religion domains in Larteh will be matched
against the rights-and-obligations (RO) sets. In the present study, the rights-and-obligations
sets may be likened to the known functions of the three major languages which co-exist in a
triglossic relationship in Larteh. In this sense, the choice of a particular language is viewed
with reference to the extent its use matches community expectations for the domain or type of
activity.
4.0 Sociolinguistic Background of the Larteh Community
Larteh is a town in the South-east of Ghana, located on the Akono Hills, the range of
hills which cross Akuapem from south-east to the north-west. It is bordered in the north by
the Mamfe-Akropong road; in the south by the Shai Hills, in the east by the towns of Abonse
and Aseseeso, and in the west by the Apopoano Hill near Dodowa. The closest neighbors are
Mamfe and Akropong where Akuapem Twi is spoken, and Dodowa and Ayikuma, where
Dangme is spoken. The Population and Housing Census of Ghana (2000) put the population
of Larteh at 8,310.This clearly does not represent the exact number of speakers because of the
presence of immigrants in the town. Yet another difficulty in arriving at a specific number
stems from the fact that there are many Leteh speakers who reside outside Larteh.
Larteh is a neighbor of language communities like the Akan and Dangme. Her closest
Guan community is Abiriw, about 10 kilometers away, where Kyerepong is spoken.
Interactions among these language groups are through inter-marriages, celebrations of
traditional festivals, trade and education. Consequently on a typical market day in Larteh, the
multilingual skills of the traders within Leteh and those from neighboring towns are utilized.
During the celebration of traditional festivals, it is common to find people from neighboring
communities in attendance to lend support. Short distances between Larteh and neighboring
towns allow school children to enroll in schools outside Larteh. This has further exposed
young school children to other languages besides Leteh, their first language. Furthermore, the
use of Twi as the language of instruction at the lower Primary level makes it mandatory for
school children in Larteh to pay attention to Twi. Contact with English and Twi has resulted
in remarkable phenomena such as borrowing, codeswitching and diglossia within the Larteh
community.
The Basel Mission established the first school in 1858 with 12 children. Presently, the
town boasts of 6 pre-schools, 7 primary schools, 7 Junior High Schools and 3 Senior High
Schools. In addition are 2 privately-owned Commercial Schools. Many children have
therefore had the opportunity to school, at least to the Junior High level, thus enhancing their
competence in Akuapem Twi which is the medium of instruction at the Basic level of
education and also, a school subject in Larteh and the surrounding Akuapem towns. The
number of schools presently, is in sharp contrast to Johnson’s report of 1973 when there were
6 Primary and 5 Middle Schools at Larteh. According to his report, Larteh had 1 Private
Commercial School and 1 government-assisted Secondary School. The increase in the
number of schools has therefore led to a corresponding increase in the use of English as a
greater number of Larterians have now been exposed to English through the school
system.The Basel Mission was the first missionary body to gain roots in Larteh, establishing
the Presbyterian Church in 1853. This was followed by the Wesleyans (Methodist) in 1879
and the Anglican Church in 1913. After the establishment of these churches, other churches
usually referred to as ‘Charismatic’ churches were also founded. Churches in Larteh may
therefore be broadly categorized into two: Orthodox Churches and Charismatic Churches.
The major orthodox churches are the Presbyterian, Methodist, and Anglican churches. The
Church of Pentecost, the Apostolic Church of Ghana, Jesus Generation Church and
Lighthouse Chapel International fall under the umbrella of Pentecostal churches in this study.
Funeral ceremonies in Larteh are communal where family members are expected to make
donations in cash and in kind to finance the funeral. A typical funeral ceremony in Larteh
covers three days, and regardless of the society being male-dominated, women play
significant roles in funeral activities.
Previous studies have recorded that in Larteh, Leteh is the first language of almost
everyone, and Akuapem Twi is spoken by almost everyone as the second language, a
phenomenon which is corroborated by the present study. A study by Stewart (1972: 83)
reports that the use of a second language in Larteh is a common mark of Guan speakers, and
that “Except in the case of Gonja, it seems that most speakers of Guan languages speak a
second language as a lingua franca.” The use of Akan as a second language by Leteh
speakers may be explained by geographical factors, first and foremost, and secondly by the
effects of national educational policies which stipulate that Akuapem Twi is used as the
medium of instruction at the Basic level of education in Larteh and surrounding towns.
Educated Larteh people speak English in addition to Leteh and Twi since English is the
language of education, administration and trade in Ghana.
Leteh is unwritten and does not possess an official orthography. The language is
mostly used in the homes, among the speakers and also as a means of communication at
traditional gatherings like funerals, festivals, marriages and naming ceremonies. Brokensha
(1966: xvi) sums up the linguistic situation in Larteh in the following words:
Three languages are in common use in Larteh- Guan, Twi and
English. Guan is generally the domestic language, children learning it
as their first language and using it in their early years. At school,
children learn English and they also have lessons in what is called
‘Vernacular’, which is in fact Twi, for Guan is not taught at any
school. Twi to some extent occupies the position of a prestige
language, as does English on other occasions, being for example used
in church.
It is common knowledge that multilingualism is the norm in Africa, and in many
communities in Africa, it is significant to find out what criteria multilingual speakers employ
when they need to choose between various languages. It is reported for instance that in
Kenya, there are more than forty languages which are spoken besides English and Kiswahili,
the official languages and the main lingua francas. An urban dweller therefore needs to
master the use of more than one language in order to communicate effectively in the various
speech domains (Kamwangamalu 2000). It is however noted that community multilingualism
does not always imply individual multilingualism, but in the case of Larteh, available data
indicate that monolinguals are almost non-existent (table 2). In table 1, languages that are
reported as being used in the community are recorded. A sample size of 418 inhabitants
which represented about 5% of the total population of 8, 310 (Ghana Population and Housing
Census, 2000) was used.
Table 1. Larteh Community Linguistic Repertoire
LANGUAGES
CLAIM
Leteh
322
Twi (Akuapem)
362
English
227
Ga
81
Dangme
29
Hausa
11
Ewe
19
French
10
Bono
1
Dagaare
2
Frafra
1
Kyerepong
6
Wala
1
Krakye
1
Gonja
1
Gwa/ Anum
1
Nzema
1
Yoruba
1
Wassa
1
Sehwi
1
USE
269
354
102
8
3
3
1
7
1
2
1
4
1
1
1
1
1
1
1
1
Data in table 1 distinguish between people who claim knowledge of Leteh from those who
actually use the language in their day-to-day activities. According to the data, there are about
20 languages spoken in Larteh. Of this number, three languages are dominant: Leteh, Twi
and English. 322 of the sample claim that Leteh is their first language, but 269 of this number
use Leteh whereas 53 of them prefer to use Twi and other languages. Out of the 418
respondents, 362 claim they can speak Akan whilst 354 of the number actually use Twi
effectively. In the case of English, 227 claim that they are able to communicate in English
whereas 102 of the number use English in some of their interactions. The scenario confirms
the multilingual nature of the community.
Individual multilingualism in Larteh is also attested as data in table 2 depict. In order to
investigate individual linguistic repertoire, the 269 respondents who indicated that Leteh was
their first language were considered. This number represents Larterians3 in this study.
Table 2. Individual Linguistic Repertoire
LANGUAGE
NUMBER
OF PERCENTAGE
SPEAKERS
REPRESENTATION
Leteh only
1
0.3%
Twi only
3
1.12%
Leteh and Twi only
265
98.51%
Leteh, Twi, English
252
93.68%
Leteh, Twi, English & other 113
42.01%
Out of the population sample, only one person is a Leteh monolingual. This is a75-year-old
woman who has had no formal education. Similarly for Twi, three people claim that it is the
only language they speak. These people had Leteh as their first language; they previously
spoke Leteh when they were children, but lost their proficiency when they migrated to live in
Northern Ghana for a greater part of their lives. As many as 265 people are bilingual in
Leteh-Twi. A little over 90% of the population sample is multilingual in the three major
languages spoken in the community. Furthermore, 113 of the population sample spoke other
languages apart from the three major languages, representing about 42% of Leteh speakers.
The data therefore substantiate a community where almost everyone is multilingual.
5.0 Language Choice in a multilingual classroom
The current Educational Language Policy of Ghana supports the use of mother tongue
in teaching at the Lower Primary level. The policy stipulates that the majority of instructional
time is spent on L1 (90% in Kindergarten and 80% in Grade 1) while time for English
gradually increases to 50% by Grade 3. The use of English as a medium of instruction is
expected to take effect from Grade 4 (National Literacy Acceleration Programme (NALAP 4)
Baseline Assessment 2009). Teaching sessions in Primary 1 of three public schools were
observed to explore how teachers manipulate the three major languages used in the
community. It was ascertained that all the pupils were bilingual in English and Twi.
Secondly, the teachers who taught in these schools were also multilingual in three or all of
these languages: English, Akan, Leteh, and Anum. Table 3 illustrates the multilingual
background of the pupils.
3
The term Larterian refers to the people of Larteh; people who consider Larteh as their hometown.
NALAP was formed to address the literacy crisis in Primary Education in Ghana.
4
Table 3. Linguistic Repertoire of Pupils
School
Presbyterian
Primary
Number on Roll
40
English
40
English, Akan
40
English, Akan, Leteh
38
Anglican Primary Methodist
Primary
38
17
38
17
38
17
35
14
Three teaching sessions were observed. The first lesson was in Environmental Science at the
Anglican Primary School, and the topic was ‘Dental Care’. The topic was mentioned in
English and then translated into Akan. The following is an excerpt of the teaching session.
Figure 1
Teacher: Everybody, let me see your teeth. What are the uses of your teeth?
Dεn na wode wo se yε?
1st Pupil: chewing
2nd pupil:mede we nam
I use for chewing meat
Teacher: Are our teeth useful then?
Class (chorus): yes!
Teacher: Today, we want to learn how we can take good care of our teeth.
Yebesua sεnea yebetumi ahwε yεn se so yiye.
Teacher: What is the first thing you do in the morning?
3rd Pupil: We wash our face, we brush our teeth.
Teacher: What do we use for brushing our teeth?
4th Pupil: duawa
stick
5th Pupil: brush ne pepsodent
and
Teacher: how many times do we have to brush our teeth in a day?
Mpεn ahe na εsε sε yetwiw yεn se dabiara?
6th Pupil: two times
Teacher: good, well done!
The teacher introduced the topic in English, and translated it into Twi. After the lesson had
been introduced in both English and Twi, the pupils answered questions using either Twi or
English, whichever they were proficient in. When the questions required one-word answers,
the pupils spoke English otherwise Twi was used. The teacher did English-Akan
codeswitching throughout the teaching session.
The second session which was observed was a mathematics lesson. Language choice
of pupils was quite different, with English language dominating in the answers that pupils
gave. The use of English dominated in the answers, because the questions did not demand
complete statements. When the teacher had introduced the topic ‘Simple Division’, she went
ahead to explain the concept in Akan using objects.
Figure 2
Teacher: Nnipa baanu kyε akutu anan a, obiara benya ahe?
When two people share four oranges, how many will each have?
st
1 Pupil: two
Teacher: Nnipa baanu kyε kwaadu asia nso ε?
And when two people share six fingers of bananas?
nd
2 Pupil: three, three
Teacher: And when three people share nine pencils?
Na nnipa baasa kyε pencil akron nso ε?
rd
3 pupil: three
The third lesson that the researcher observed was in Religious and Moral Education where
the teacher taught the ‘Creation Story’ from the Holy Bible. The Bible was read in English,
after which the teacher narrated the account in Twi. All the questions were asked in Twi, and
the responses from pupils were mostly in Twi. After the lesson, there was a memory verse
taught in English. Below is an excerpt of the class session.
Figure 3
Teacher: Dɛn na Onyankopon bɔ dii kan?
‘What did God create first?’
Pupil 1: star, nsoroma
Teacher: Nna ahe na ɔde bɔɔ biribiara?
‘How many days did He use in creating everything?’
Pupils 2: Nnansa
‘Three days’
Teacher: Dabi
No
Pupil 3: 6 days
Teacher: good!
The teacher narrated the story in Twi after reading from the English Holy Bible, because she
intimated that biblical English was sometimes difficult to understand. She could not read the
Twi bible so she read in English and then explained in Twi. Pupils gave some one-word
answers in English especially where it was a loan word or lexical items which were better
known in English. In most of the instances, the answers were in Twi. The memory verse was
taught in English, because it gave the pupils a sense of pride to recite Bible verses in English.
During school Open Days and Children’s Day in the Presbyterian Church for instance, school
children were made to recite English Bible verses to the admiration of unlettered parents. For
the parents of these school children, the ability to recite Bible verses in English was evidence
of literacy.
The three preceding teaching scenarios present language choices in the Primary 1
classroom. In each case, the teacher’s language choice may be described as unmarked. The
bilingual strategy adopted by the teacher is what is expected from a Primary 1 teacher as per
the Government of Ghana’s educational policy on language. The teacher’s speech was
therefore characterized by codeswitching. The type of lesson also determined the teacher’s
language choice. In this case, the appropriate language had to be selected in order to achieve
teaching and learning objectives as seen in the Religious and Moral Education lesson. On the
other hand, the pupils’ choices stemmed from their competency levels in the two prescribed
languages of instruction. The institutional factor may be likened to Johnson (1975)’s
situational rule which stipulates that it is a situation type that determines the choice of
language. In the classroom situation therefore, the language (s) is/are prescribed, and the
teacher does not have much control over the language of instruction.
Applying the Markedness Model, we can assert that the teacher’s unmarked choice of
language was dictated by the Educational Language Policy of Ghana. The choice was
therefore normal and matched community expectations. For the pupils who may have had
very little knowledge or none at all about language policy, their English-Twi bilingualism
was exploited. The language choices that pupils made in answering questions indicated their
proficiency levels in the two languages: English and Twi. Again, one would describe their
language choices as umarked since they were expected.
It must be noted that Leteh, the first language of the majority of the pupils was not
used in any of the teachings sessions. If any of the speech participants had spoken Leteh, that
would have been a marked choice, and would have carried ‘extra social meaning’ (Kieswetter
1995: 25). This pattern of language use in education is similar to what pertains in some
multilingual communities in Africa. For instance, Ncoko et al (2000: 239, 231) reporting on
the benefits of codeswitching for teacher education recommended that, in the multilingual
schools in South Africa, codeswitching could be used as a teaching strategy since it had
communicative functions such as translation, checking comprehension, giving instruction and
clarification. Similar to observations made in the present study, Ncoko et al also found that
learners utilize their linguistic ability and resources to control their conversations according
to content and circumstances.
Table 4 summarizes the three scenarios of teaching sessions. In each session, the
language which predominates is listed first.
Table 4. Patterns of Language Choice in Primary 1 Classroom
Domain
Setting
Speaker Addressee Subject
Topic
Education School
Teacher
Pupils
Integrated
Dental
Science
Care
Mathematics Simple
Division
Religious
Creation
and Moral
Education
Language
English/Akan
English/Akan
Twi/English
6.0 Language Choice at Funerals.
Funerals at Larteh are public, and are characterized by the attendance of many
mourners and sympathizers. As sympathizers may be of diverse language backgrounds,
language choice is made with care so that communication goals are realized. To a large
extent, language choice is dependent on the acceptable social roles of each language, the
rights-and-obligations (RO) sets (Myers-Scotton 1993: 84). For each sub-activity of the
funeral therefore, participants expected a particular language choice. In Larteh, funeral
activities normally span three days, Friday-Sunday. The dead body is laid in state from Friday
late afternoon to the morning of the Saturday. In the evening of the Friday, close relatives of
the deceased gather in the family house to keep wake. The tradition is that any sympathizer
who comes around to greet the family is given an update or a briefing on the cause of death,
which is usually done in Leteh first, and then summarized in Twi.
On the following day, the family head performs a rite to select a successor for the dead
person. The gathering was usually made up of close family members who are mostly
Larterians. It is a short ceremony during which Leteh is largely spoken. The following is an
extract of a funeral session where the deceased lived and worked in Kumasi, a town in the
Ashanti Region of Ghana. On this occasion, Leteh (L) is spoken and then interpreted into Twi
(T).
Figure 4
(Sympathizers enter the funeral grounds to greet the bereaved family; the family spokesman
introduces them).
Family Spokesman: Agoo!
Attention!
Ɛsε ne debiakε a twu Kumase pεε. Amo gyi Papa… esumitese. Amo debεboaokunafo a, ne
ahurε mo kuru.(L)
Adɔfonom, nnipa yi a wogu so rekyia yi fi Kumase pεε. Wɔyε Agya…adwumayεfo. Wɔaba
rebεboa okunafo no asie ne kunu.(T)
‘The people who are greeting used to be the co-workers of Mr…. They have come all the way
from Kumasi to help the widow bury her late husband.’ (E)
During the funeral, various announcements are made. These announcements are meant to
direct visitors to venues of funeral activities, and also to inform the gathering of donors.
Announcements are made in all the three major languages: Leteh, Twi and English due to the
diverse language backgrounds of mourners and sympathizers. A scenario is illustrated in
figure 5.
Figure 5
Family Spokesman: Ɛne debεna asɔretsa nte. Ene ntwu asiei mbε a, εne bebunkyi bε mfe
ara (L)
Yεrebεkɔ asɔredan mu. Se yefi amusiei a, yεbεsan abehyia wɔ ha bio. (T)
We shall go to the chapel from here. From the cemetery, we shall come
back here. (E)
In the first instance (figure 4), the family spokesman used Leteh and Twi. He interpreted his
statements into Twi for the benefit of the sympathizers who had arrived from Kumasi. This
scenario differs from the one during which only Leteh was spoken, the selection/election of a
successor. In figure 5, all the three languages are used in order to reach a larger audience. The
family spokesman assumes the role of host and ensures that his ‘guests’ do not lose their way
whilst they are in town. It is also important to mention the names of donors so that they are
well appreciated. The spokesman therefore uses Leteh, Twi and English.An interaction of
marked and unmarked language choices was observed. Although it was expected that during
a traditional ceremony, Leteh will be the unmarked choice, this was not always the case due
to the diverse language backgrounds of speech participants. The use of English in the
traditional domain does not match community expectations, and as a result the choice may be
described as marked so far as domain type was concerned. Nevertheless, Leteh, the unmarked
choice was used in performing some exclusive traditional rites. The mourners and
sympathizers used various languages among themselves, according to their linguistic
repertoires. In selecting a language to address the mourners and sympathizers, the speaker
always considered their linguistic repertoires; he therefore used all the three major languages.
The alternate use of unmarked and marked languages confirms the assertion that, within the
markedness model, code choices are intentional in that they are made to achieve specific
social ends. In this sense, the speaker will choose one language variety over another, because
it has more benefits than costs (Myers-Scotton 1998:19). The repertoire rule as proposed by
Johnson (1975) dictated marked language choices to a large extent in the traditional domain.
Table 5 is a summary of patterns of language choice during the various sub-activities of a
funeral in Larteh. For each of the activities, the predominant language is listed first.
Table 5. Patterns of Language Choice at Funerals
Domain
Speaker
Addressee
Setting
Tradition
Family
Mourners and Funeral
Spokesman
Sympathizers
Activity/Topic
Introduction
sympathizers
Language
of Leteh,
Twi
Reporting cause of Leteh,
death
Twi
Electing a successor Leteh
Announcements
Leteh,
Twi,
English
7.0 Language Choice in the Church
For the purposes of the present study, churches at Larteh were categorized into two:
Orthodox and Charismatic churches. The classification is necessary because mode of worship
in these two types of churches differ significantly, a disparity which has the potential of
influencing language choice.In the category of Orthodox churches, the Presbyterian and the
Catholic churches were observed. During a normal Sunday morning service at the
Presbyterian Church, the language for worship was predominantly Twi. References of Bible
Readings were however made in English. There were a few cases of Akan-English
codeswitching in sermons. Leteh was not used in the course of the church service. This
differed from special occasions, for example, a funeral service where the congregants had
various language backgrounds. During such funeral services, the sermon was delivered in
Twi and then a summary given in English.
The following field notes which were taken on January 22, 2012 illustrate language
choice in a normal Sunday worship service in the Presbyterian Church. Call to worship was
done in Twi by the Senior Minister. A prayer was said in Twi, and hymns were also sung in
Twi. When a hymn was introduced, the reference was first mentioned in Twi, then in English.
Bible readings were done in Twi, and similarly the references were said in both Twi and
English. The choir and the singing band sang in Twi. The congregants used Twi hymn books.
The Session Clerk (church secretary) read the announcements in Twi. The Senior Minister
said the closing prayer and pronounced the benediction in Twi. After the church service,
interactions among the congregants were in both Leteh and Twi.
Language choice in the Presbyterian Church may be accounted for by its history.
Christianity was introduced in Larteh by the Basel Mission whose basic policy concerning
missionary work was the utilization of Ghanaian languages. Missionary work started at
Akropong-Akuapem; Akuapem Twi was therefore adopted as the literary standard for use in
the churches where Twi is spoken. Other factors accounting for the predominant use of Twi
over Leteh in the church are that, many Christian ministers who are posted to Larteh to head
the orthodox churches are non-Larterians. In addition, all Christian materials are printed in
Twi and English, and none in Leteh. It was however gathered that whenever the chief of the
town attended church service, and he was given an opportunity to address the congregation,
he spoke Leteh.
Language choice in the orthodox churches, typified by the Presbyterian Church is
governed by factors like domain-type and occasionally by the linguistic repertoire of speech
participants. In the first place, the unmarked choice was Twi, because it is the language of
Christian religion in Larteh and the surrounding Akuapem towns in Ghana. Secondly, the
varied repertoires represented during special services such as a funeral service required the
use of English in addition to Twi. The choice of English, an unmarked choice, is therefore
said to carry additional meaning. Finally, the choice of Leteh over Twi by traditional leaders
who visited the church occasionally may be described as figurative. By using Leteh, a marked
choice, the chief built solidarity with his subjects. Johnson’s rules bordering on situation,
linguistic repertoire and metaphoric are applicable to language choice in the orthodox
churches in Larteh.
Language choice in the charismatic churches was quite different. The following are
field notes taken at the Lighthouse Chapel International Church branch at Larteh on February
5, 2012. The church service began with a prayer led by the Assistant Pastor. The assistant
Pastor whose first language is Ewe used English. During the prayer time, members prayed
mostly in English. This was followed by a period of singing. The songs were in Twi. The
song leader did English-Twi codeswitching to urge the congregants to sing and praise God.
Congregants were heard praying in both English and Twi. When it was time for church
members to share testimonies of God’s intervention in their lives, they used Twi with a few
English phrases such as ‘Praise the Lord’; ‘My sister encouraged us; everybody encouraged
us’. Although the two speakers were Larterians, they spoke Twi mainly. The sermon was
characterized by Twi-English codeswitching (Figure 6). Both the Twi and English bibles
were read. The Pastor welcomed visitors using Twi and English. At the close of the church
service, members recited a prayer (grace- a scriptural verse which has been memorized by
members) to one another in English whilst they shook hands.
Figure 6 Codeswitching in Sermon
Pastor: I told you that sε wofa bible a, baabi a εyε kɔkɔɔ no, Yesu na ɔkaeε…
‘I told you that when you pick a bible, the red parts are the direct sayings
of Jesus…’
Enti this year, we have all decided that this year we want to make it. Okristoni,
nneεma bebree wɔ hɔ a εsε sε yεyε.
‘So this year, we have all decided that this year, we want to make it. Christian,
there are many things which we ought to do.’
Unlike the Presbyterian Church, the Lighthouse Congregation had membership which was
younger, mostly between ages 10-40. The members were mostly pupils and students which
accounts for the comparatively more use of English. The predominant use of English also
stems from its foundation; the founders of the church encouraged the use of English right
from its establishment.
In the Charismatic church, there was a combinatory use of unmarked and marked
choices. From the perspective of the visitor who has no knowledge of the RO sets of the
church, the expected language choice was Akuapem Twi. However, the church members who
shared the RO sets of the church would consider Akuapem Twi as a marked choice as
opposed to the expectations in the Orthodox Church. Although the majority of the
worshippers were Larterians, Leteh was not used. Table 5 summarizes language choices that
are made in the churches.
Table 5a. Patterns of Language Choice in the Presbyterian Church, Larteh
Domain
Speaker
Addressee
Setting
Activity
Language
Christian
Pastor
Worshippers Presbyterian Call to Worship Twi
Religion
Church
Church
Prayer
Twi
Leader,
Church
members
Congregation,
Singing
groups
Church
members
Pastor
Church
Secretary
Pastor
Singing
Twi
Bible Reading
Twi
Sermon
Twi
Announcements Twi
Benediction
Table 5b. Language Choice at the Lighthouse Chapel, Larteh
Domain
Speaker
Addressee
Setting
Activity
Christian
Assistant
Worshippers Lighthouse Preparatory
Religion
Pastor
Chapel
Prayer
Singing
Senior
Sermon
Pastor
Church
Bible Reading
members
Church
Testimonies
members
Senior
Announcements
Pastor
Worshippers
Sharing
of
grace
Twi
Language
English
Twi
Twi/English
Twi/English
Twi/English
Twi/English
English
8.0 Summary and Conclusion
The paper set out to explore factors that dictate language choices made by
multilingual speakers in Larteh, a multilingual community. The study was conducted within
the framework of Markedness Model (Myers-Scotton 1993; 1998). Three domains that were
investigated are education, tradition and Christian religion. The results of this study are
founded on data gathered from a linguistic survey, interviews and participant observation.
Research done on language choice in multilingual communities have established that among
many considerations that multilingual speakers make in selecting a language for
communication are the domain of language use, topic/activity and the language background
of speech participants (Kamwangamalu 2000). Within the framework of markedness, the
social and psychological associations that speakers attach to languages inform choices that
are made. A language choice may therefore be perceived as unmarked or marked, depending
on the extent to which it matches community expectations. In this study, community
expectations are aligned with the set functions of the three major languages, typical of a
triglossic community. In this paper, these expectations, which are shared by members of the
Leteh community represent Myers-Scotton’s rights-and-obligations sets. Johnson (1975)’s set
of rules for predicting language choice were also applied to the data where relevant.
This study has shown that in addition to factors of language choice documented in the
literature, national language policies play significant roles in determining language spoken in
some speech domains.
In the educational domain, it has been found that the teachers’ choice of both English
and Twi in teaching Primary 1 is largely dictated by the Ministry of Education’s Policy on
Language. On the part of pupils, their proficiency level in either of the two languages
governed their language choice during classroom interactions. The RO set in the educational
domain, therefore accounts for language choices. Language choice in the domain of tradition,
for example at funerals, was based on the linguistic repertoire of addressees. A combination
of marked and unmarked choices was observed. Accordingly it came out that all the three
major languages: Leteh, Akuapem Twi and English were employed during funeral activities.
Finally in the religious domain, whereas Akuapem Twi was the unmarked choice in the
orthodox churches, it was considered as the marked choice in the Charismatic churches. In
the Charismatic Church, the history of its establishment was such that English was accepted
as the unmarked language choice. Church services were also characterized by English/
Akuapem Twi codeswitching. Language choice in the Orthodox Church was largely
governed by type of domain, whereas in the Charismatic Church, it was a combination of
factors.
The functional distribution of the three languages: Leteh, Twi and English as reported
by earlier researches, Brokensha (1966); Johnson (1973, 1975) has changed with regard to
the use of English. In the educational domain for instance, English is combined with Twi as
languages of instruction as directed by the Language Policy on Education at the Basic Level.
Again, the introduction of English in traditional ceremonies such as funerals is also
noteworthy. This is contrary to Johnson’s report (1973: 122) that Leteh is the principal
language in traditional ceremonies. Similarly, with the founding of Charismatic Churches,
English has been introduced as a key language in the Christian Religious domain. The
introduction of English in some domains, as well as the comparative increase of its use may
be accounted for by a corresponding growth in education in the Larteh community.
i.
In conclusion, I make the following assertions:
Social change in Larteh over the past three decades has influenced factors that dictate
language choice
ii.
Domains where English is used have increased to the detriment of Leteh, thus posing
threats of endangerment
iii.
In the face of multifaceted changes that have taken place in Larteh, which have also
affected language use in the community, Johnson’s (1975) assertion of ‘stable
triglossia’ is arguable.
iv.
In view of the above, it is recommended that conclusions that are drawn from studies
on language use in multilingual communities need to be reviewed periodically in line
with global and local changes
References
Bodomo, A., Anderson, J. & Dzahene-Quashie, J. (2009). A kente of many colours.
Multilingualism as a complex ecology of language shift in Ghana. Sociological Studies, 3
(3), 357-379.
Brokensha, D. (1966). Social change at Larteh, Ghana. Oxford: Clarendon Press
Ghana Statistical Service. (2002). 2000 Population and housing census: Summary report of
final results. Accra: Ghana Statistical Service.
Johnson, B. C. (1973). Language use at Larteh, Ghana. A sociolinguistic study of a bilingual
community. (Unpublished doctoral thesis). Northwestern University,
Johnson, B. C. (1975). Stable triglossia at Larteh. Patterns in language, culture and society:
Sub-Saharan Africa. OSU WPL 19. 93-102
Kamwangamalu, N. M. (2000). Languages in contact. In V. Webb and Kembo-Sure (eds.).
African Voices. An Introduction to the Languages and Linguistics of Africa. Southern Africa:
Oxford University Press
Leherr K. (2009). National Literacy Acceleration Program (NALAP) Baseline Assessment
Document. Education Development Centre. USAID/Ministry of Education, Ghana
Myers-Scotton, C. (1993). Social motivations for codeswitching: Evidence from Africa.
Oxford: Clarendon Press
Myers-Scotton, Carol. 1998. A theoretical introduction to the markedness model. In C.
Myers-Scotton ((ed.) Codes and Consequences: Choosing Linguistic Varieties. New York:
Oxford University Press
Ncoko, S. O. S., R. Osman and K. Cockroft. 2000. Codeswitching among Multilingual
Learners in Primary Schools in South Africa: An exploratory Study. InternationalJournal of
Bilingual Education and Bilingualism 3(4): 225-241
Kieswetter, A. 1997. Code-switching amongst African High School Pupils. University of
Witwatersrand Occasional Papers in African Linguistics
Stewart, J. M. 1972. Language. In David Brokensha (ed.). Akwapim Handbook. Accra: Ghana
Publishing Press
Yakubu, A., Matu, M. P., Ongarora, D. (2012). Language use and choice: A case study of
Kinubi in Kibera, Kenya. International Journal of Humanities and Social Science 2 (4), 99104.